Jump to content

What Is Wrong With The Psr.


63 replies to this topic

#41 Pendaelose

    Member

  • PipPipPip
  • Elite Founder
  • Elite Founder
  • 69 posts

Posted 30 January 2016 - 12:03 PM

Win/Loss ratio is still the best way to go.

All matches are symmetrical, so for you to win, someone has to lose. The Average Win/Loss Ratio is always 1, and if you are correctly matched against other players of equal skill your win/lose will be 1 (or VERY close to it)

If for any reason your Win/Loss ratio is significantly greater than 1 it can only mean one of 3 possible things.
  • Your Ranking is inaccurate and it is causing the Match Maker to match you against weaker players. If this is true we need a more reliable metric for Ranking than our current PSR
  • Your Ranking is accurate and the Match Maker is matching you against players it knows are weaker. If this is true we need to decide if balanced matches are worth the longer wait time.
  • Both systems are borked and all matching is essentially random. If Match making is random and your being matched against the full range of players then your Win/Loss is simply a reliable metric of how much better you are than the average player.
And of course, all of this is true in reverse for weaker players who might be paired with stronger players.

#42 Davers

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,886 posts
  • Facebook: Link
  • LocationCanada

Posted 30 January 2016 - 12:19 PM

People keep saying that PSR doesn't work. Well, where are the top players who are stuck in Tier 3-4? Where are the Tier 1 guys who die to LRMs every match? Where are the actual examples of PSR not working?

#43 Torgun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,598 posts

Posted 30 January 2016 - 12:37 PM

View PostDavers, on 30 January 2016 - 12:19 PM, said:

People keep saying that PSR doesn't work. Well, where are the top players who are stuck in Tier 3-4? Where are the Tier 1 guys who die to LRMs every match? Where are the actual examples of PSR not working?


Since we can't see others' tier in-game, your questions are actually impossible to answer. What we do know is that if you are an average player doing really average, PSR will slowly carry you higher beyond middle of tier 3. And the more matches you play, the faster you advance. It's simply how PSR is (badly) designed. Everyone gets ranked higher than they should the more matches they play, unless you top out at tier 1. It just makes no sense and the only reason I can see why they designed it as such is to stroke bad players' egos so they seem better than what they actually are. It's only when they do a fair bit worse than average that the PSR ranking will finally stop rising.

#44 C I L L I P U D D I

    Member

  • PipPipPipPipPip
  • The Butcher
  • The Butcher
  • 188 posts

Posted 30 January 2016 - 12:55 PM

What does it even matter? If the PSR is there just to stroke egos, why do you care? You should want to play against better players, it's the only way you get better. If you don't want to get better then it doesn't matter who you play against because you're just there to have fun. The problem is that people who are wildly average want to club seals and since they've gotten placed in Tier 3 they get put up against T1-2's and they become the seals. When I queue I don't care who is on my side or the opponents, in solo. I look at the teams to see if I know anyone and throw them a o7 or Hi "said player" but that's it. It has no bearing on how I play or the steps I take to try to win. Now in group queue, it's different, because when you start seeing some of the higher tier teams run together you need to learn how they play and where they like to push on the map. It still has little to no bearing on my play though, as far as my ability to pilot a mech.

#45 Davers

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,886 posts
  • Facebook: Link
  • LocationCanada

Posted 30 January 2016 - 12:59 PM

View PostTorgun, on 30 January 2016 - 12:37 PM, said:


Since we can't see others' tier in-game, your questions are actually impossible to answer. What we do know is that if you are an average player doing really average, PSR will slowly carry you higher beyond middle of tier 3. And the more matches you play, the faster you advance. It's simply how PSR is (badly) designed. Everyone gets ranked higher than they should the more matches they play, unless you top out at tier 1. It just makes no sense and the only reason I can see why they designed it as such is to stroke bad players' egos so they seem better than what they actually are. It's only when they do a fair bit worse than average that the PSR ranking will finally stop rising.

Hence why some 'tweaking' is coming. But the OP's opinion that the system is fatally flawed because it is based on W/L primarily doesn't really hold up.

#46 Torgun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,598 posts

Posted 30 January 2016 - 01:01 PM

View PostCillipuddi, on 30 January 2016 - 12:55 PM, said:

What does it even matter? If the PSR is there just to stroke egos, why do you care? You should want to play against better players, it's the only way you get better. If you don't want to get better then it doesn't matter who you play against because you're just there to have fun. The problem is that people who are wildly average want to club seals and since they've gotten placed in Tier 3 they get put up against T1-2's and they become the seals. When I queue I don't care who is on my side or the opponents, in solo. I look at the teams to see if I know anyone and throw them a o7 or Hi "said player" but that's it. It has no bearing on how I play or the steps I take to try to win. Now in group queue, it's different, because when you start seeing some of the higher tier teams run together you need to learn how they play and where they like to push on the map. It still has little to no bearing on my play though, as far as my ability to pilot a mech.


I really need to explain why it matters if the tier rankings are badly done, since all matchmaking is done using said tiers to make competitive matches? Seriously?

#47 C I L L I P U D D I

    Member

  • PipPipPipPipPip
  • The Butcher
  • The Butcher
  • 188 posts

Posted 30 January 2016 - 01:07 PM

What does it matter though? Honestly. You play against the players that are selected and you either win or lose. I've lost to bad teams and I've beaten good teams. Whether you do well individually as a pilot has everything to do with you and not MM. MM will never increase your piloting skill and will never get you make better decisions. The only constant in all your matches is you.

#48 Torgun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,598 posts

Posted 30 January 2016 - 01:11 PM

View PostCillipuddi, on 30 January 2016 - 01:07 PM, said:

What does it matter though? Honestly. You play against the players that are selected and you either win or lose. I've lost to bad teams and I've beaten good teams. Whether you do well individually as a pilot has everything to do with you and not MM. MM will never increase your piloting skill and will never get you make better decisions. The only constant in all your matches is you.


Basically you're saying you don't care about matchmaking at all, since if it's all messed up doesn't bother you. At which point why do we even have tiers if there's no use of said tiers to make competitive matches?

#49 DAYLEET

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 4,316 posts
  • LocationLinoleum.

Posted 30 January 2016 - 01:14 PM

View PostAshvins, on 30 January 2016 - 12:51 AM, said:

So in the Town Hall meeting Russ mentioned that the PSR (Pilot Skill Rating) needed a few "tweeks". Well to be blunt about it, Win or Loss factors too much into the PSR. You can have a Good Match (not Great just good) like this one and loose. and have your PSR stay equal or Go down.Posted Image

On the other hand you can have a total garbage match and have your PSR go UP as shown Here.Posted Image

I intentionally sabotaged myself in the 2nd match shown here to prove the point. Terrible damage, but a win and my PSR goes up. This should not happen. The win/loss factor for the PRS is what is out of wack. My PSR should have gone UP in the first match thanks to a good performance. In the 2nd it should have stayed the same or gone down due to the poor performance.

Any given pilot can only do so much to influence the win. In a situation where your team is not performing well yet you do, you should not be penalized through your PSR. Your already penalized in the pocket through the loss and lower C-bill earning, penalizing the PSR on top of it is adding insult to injury.

4X the damage yet 1/2 the C-bills is enough of a penalty for a Loss. Stop the reduction of our PSR because our teammates could not carry their weight. And most definately Stop the increase of PSR for terrible performance on a win.

Luck should have nothing to do with the PSR and with the way it is a terrible player can get lucky by having a good team and increase his PSR. Increasing his rank and then hindering people of higher rank with his poor play when he no longer get's lucky.

Let's look at the XP gained as well (BROKEN) 4X the damage, a kill a KMDD, Comp Dest, and 2/3 the XP of the win with terrible stats. I understand the C-bills being less for the loss but XP and PSR should not be affected by the Win or loss in any given match.

So there you go WHAT IS WRONG WITH PSR.

First: No, 400 damage and 1kill should not raise your psr in a loss. In my opinion, nnot how it is in game, you would need to do some serious carrying in a loss like 800 damage and 4+ kills to gain psr, no middle ground on a loss.

You only take damage into account as if anyone who do 400 damage did good, you could have spread it across 12 mech for all we know and be largely useless for your team. Your win also had a well placed uav which is very useful for a team who isnt completely inept and maybe your 100 damage was better placed and damage soaking better timed. You got double cbills and xp on a win, even though its conq and that pays more, you manage to be a better team player on your win than on your loss.

View PostTorgun, on 30 January 2016 - 01:11 PM, said:


Basically you're saying you don't care about matchmaking at all, since if it's all messed up doesn't bother you. At which point why do we even have tiers if there's no use of said tiers to make competitive matches?

Take the new players out of the farming reach from the vets.

View PostEl Bandito, on 30 January 2016 - 07:32 AM, said:

T1s arguably have the toughest time to reach the top of the leaderboard. Leaderboards is chiefly about grind and luck, and exploiting group queue by feeding one of their unit members kills.

Back when i cared and participated, the top of the ladder was always occupied by solo players who were very good players and put a lot of hours into the game to get freak games, some even bought 2 of the same hero to grind faster. Group feeding kills, ive heard that story, if it wasnt an urban myth you can bet your *** there would be a riot on the forum.

Edited by DAYLEET, 30 January 2016 - 01:28 PM.


#50 Davers

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,886 posts
  • Facebook: Link
  • LocationCanada

Posted 30 January 2016 - 01:15 PM

View PostTorgun, on 30 January 2016 - 01:11 PM, said:


Basically you're saying you don't care about matchmaking at all, since if it's all messed up doesn't bother you. At which point why do we even have tiers if there's no use of said tiers to make competitive matches?

If you are tier 3 there isn't much matchmaking happening. Some matches you are against Tier 1-3, others Tier 2-5. Until tiers get more restricted, it only means the tier 1 guys don't get to stomp the 4-5's.

#51 C I L L I P U D D I

    Member

  • PipPipPipPipPip
  • The Butcher
  • The Butcher
  • 188 posts

Posted 30 January 2016 - 01:20 PM

It doesn't bother me. I wouldn't care if I got matched up against LORD, EMP, CSJ, or SJR every single match, in fact I would prefer it. I know that I'm going to get hammered down and lose 99% of the time against them but I know I'm getting better every time I play them. I don't go into matches going "Thanks MM", I go "Ok, last time you played these guys you did 300 damage, let's see if you can do 400 this time and maybe get a kill or 2". And you know what, my game play has improved drastically. So when I'm not matched up against them, I can just dumpster regular units and players. At some point you have to stop blaming everything else for where you're at in life and even in a video game and start looking at how you're approaching things retrospectively.

#52 Torgun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,598 posts

Posted 30 January 2016 - 01:23 PM

View PostDavers, on 30 January 2016 - 01:15 PM, said:

If you are tier 3 there isn't much matchmaking happening. Some matches you are against Tier 1-3, others Tier 2-5. Until tiers get more restricted, it only means the tier 1 guys don't get to stomp the 4-5's.


Tier 3 can end up with tier 1-5 in the match, but matchmaker will try to (and clearly sometimes fail) to even up the teams by not having all tier 1 in one team and all tier 3 in the other for example. But since the tiers right now are messed up due to you being able to climb a tier or two higher than your actual ability by simply playing tons of matches, even matches that on paper look great could in fact be quite bad. They really have to rethink how and when you go up/down in ranking.

View PostCillipuddi, on 30 January 2016 - 01:20 PM, said:

It doesn't bother me. I wouldn't care if I got matched up against LORD, EMP, CSJ, or SJR every single match, in fact I would prefer it. I know that I'm going to get hammered down and lose 99% of the time against them but I know I'm getting better every time I play them. I don't go into matches going "Thanks MM", I go "Ok, last time you played these guys you did 300 damage, let's see if you can do 400 this time and maybe get a kill or 2". And you know what, my game play has improved drastically. So when I'm not matched up against them, I can just dumpster regular units and players. At some point you have to stop blaming everything else for where you're at in life and even in a video game and start looking at how you're approaching things retrospectively.


So you don't care about matchmaking at all, and we're here in a thread talking about how tiers are made which are used for matchmaking? Basically we have nothing to talk about then. Good to know.

#53 C I L L I P U D D I

    Member

  • PipPipPipPipPip
  • The Butcher
  • The Butcher
  • 188 posts

Posted 30 January 2016 - 01:36 PM

No I asked you why do you care if PSR is just an ego stroke. To which you have never replied nor have you really made any suggestions that would improve the system.

So I will ask again... Why do you care about PSR, if it's just an ego stroke?

Splitting up the teams by Tier, evenly, isn't going to fix anything. Say you have 24 Tier 3 players in a match, that are all the exact same PSR. I mean to the decimal the same. One team 12-0's the other side. Whose fault is it then? MM? PSR?

Another example, you have 12 Tier 1's and 12 Tier 3's fight each other. The Tier 3's win 12-2. Does this mean MM is broken or those Tier 3's don't deserve to be where they are? No, it means they played better than those set of players on a set map for one game. I can bet you this though, a lot of those Tier 1's are sitting at their desk thinking about what they can do better next game, while the Tier 3's are soaking up the win thinking their god's gift to MWO.

They play again, exact same teams, exact same map, and exact same mechs. Tier 1's roll 12-1. What do you see in chat "ThanKs MM". Only difference is some of the Tier 1's understand that each game, no matter the similarities, is different. And every move you make and every shot you place on a mech actually matters. While the Tier 3 guys are too caught up and how bad MM screwed them over.

#54 Appogee

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 10,967 posts
  • LocationOn planet Tukayyid, celebrating victory

Posted 30 January 2016 - 01:37 PM

I'd be fine with win/loss being such a large factor in PSR ... if the teams were evenly matched.

But the matchmaker rarely seems to put together an even match.

Look at this recent match of mine. I'm Tier 2, and I'm expected to win in a team where 5 guys can't even do 100 damage...?

Posted Image

Noone can actually win when you've got this level of derp on your team. But my PSR will decline unless I manage to salvage 250 match score in the few short minutes before my team derps into oblivion.

God help me when I'm levelling Mechs, or playing lesser chassis. GGclose..

Edited by Appogee, 30 January 2016 - 01:41 PM.


#55 InspectorG

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Boombox
  • The Boombox
  • 4,469 posts
  • LocationCleveland, Ohio

Posted 30 January 2016 - 01:46 PM

ELO/PSR/WHATEVS wont mean crap until the objectives and 'infotech-should we get it' are hammered out.

Currently, MWO = damage done efficiently. Team that gets a 3 mech lead first has a nice advantage and 'winning' players should, in theory, facilitate that event better.

PSR is...ok. Im not expecting much until objectives/maps/balance is better.

#56 Torgun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,598 posts

Posted 30 January 2016 - 01:47 PM

View PostCillipuddi, on 30 January 2016 - 01:36 PM, said:

No I asked you why do you care if PSR is just an ego stroke. To which you have never replied nor have you really made any suggestions that would improve the system.

So I will ask again... Why do you care about PSR, if it's just an ego stroke?

Splitting up the teams by Tier, evenly, isn't going to fix anything. Say you have 24 Tier 3 players in a match, that are all the exact same PSR. I mean to the decimal the same. One team 12-0's the other side. Whose fault is it then? MM? PSR?

Another example, you have 12 Tier 1's and 12 Tier 3's fight each other. The Tier 3's win 12-2. Does this mean MM is broken or those Tier 3's don't deserve to be where they are? No, it means they played better than those set of players on a set map for one game. I can bet you this though, a lot of those Tier 1's are sitting at their desk thinking about what they can do better next game, while the Tier 3's are soaking up the win thinking their god's gift to MWO.

They play again, exact same teams, exact same map, and exact same mechs. Tier 1's roll 12-1. What do you see in chat "ThanKs MM". Only difference is some of the Tier 1's understand that each game, no matter the similarities, is different. And every move you make and every shot you place on a mech actually matters. While the Tier 3 guys are too caught up and how bad MM screwed them over.


Seriously man, correct tiers matters because matches are made using those tiers to make the chance of even matches as high as possible. And setting the tiers up just to stroke egos will make the matches even more uneven on purpose by design by the devs and that's simply stupid.

And if we try to have some common sense here, the team that have the clear majority of better players will undoubtedly have a higher chance of winning the match. Yes sometimes they lose to, but let them play each other 10 times, who do you think will come out victorious in the end?

And yes you clearly said it doesn't matter to you who you face or who you play with, so any effort to make decent matchmaking does in fact not matter to you because that is what you actually said.

Edited by Torgun, 30 January 2016 - 01:51 PM.


#57 C I L L I P U D D I

    Member

  • PipPipPipPipPip
  • The Butcher
  • The Butcher
  • 188 posts

Posted 30 January 2016 - 01:56 PM

What I've been saying is about as common sense as it gets. Why focus on who is on your team or the enemy, when you can be focused on how you can be a better pilot. No amount of changes to PSR will change that.

League of Legends has the same issue, ELO Hell as it's referred too. There is no actual Elo Hell. It's just players blaming other players and MM for them losing and not looking at their own mistakes as maybe a reason that lead to them losing.

My solution or suggestion, is to just get rid of the whole system and you get who ever you get. Good or Bad, deal with it.

Stand up and be counted or lay down and be mounted.

#58 Torgun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,598 posts

Posted 30 January 2016 - 02:04 PM

View PostCillipuddi, on 30 January 2016 - 01:56 PM, said:

What I've been saying is about as common sense as it gets. Why focus on who is on your team or the enemy, when you can be focused on how you can be a better pilot. No amount of changes to PSR will change that.

League of Legends has the same issue, ELO Hell as it's referred too. There is no actual Elo Hell. It's just players blaming other players and MM for them losing and not looking at their own mistakes as maybe a reason that lead to them losing.

My solution or suggestion, is to just get rid of the whole system and you get who ever you get. Good or Bad, deal with it.

Stand up and be counted or lay down and be mounted.


Yeah let's make one big queue and have 12-mans play pugs too because matchmaking doesn't matter. Well so my initial point was correct, you think we should have no matchmaking, in a thread discussing about how to make matchmaking better. We have nothing to discuss, moving on.

#59 C I L L I P U D D I

    Member

  • PipPipPipPipPip
  • The Butcher
  • The Butcher
  • 188 posts

Posted 30 January 2016 - 02:08 PM

I never said don't separate solo queue and group queue. I figured that was a fairly obvious thing that's needed.

Edited by Cillipuddi, 30 January 2016 - 02:09 PM.


#60 Ashvins

    Member

  • PipPipPipPipPip
  • Bad Company
  • Bad Company
  • 174 posts

Posted 30 January 2016 - 05:32 PM

Look I started this thread to get people to discuss how the make the PSR better. I pointed out that unless your carry your entire team with a 550+ damage match on a loss your PSR will stay the same. If your in the 250-340 match score it stays the same 450 or so and up even a loss will gain you PSR.

It's less about this than the player who has a 20 match score and is on the winning team and has his PSR go up. That fact is what makes the PSR broken. I'd like to see it more representative of your overall performance, win or loose. Winning and doing little to nothing, gaining you PSR is just wrong. Kicking *** as much as is possible and loosing , reducing your PSR again is wrong.

What I'm I'm saying here is if I have a horrible match in an Atlas and do under 100 damage my PSR should go DOWN even if my team wins.

Again if I do 300-400 damage in an Atlas before being the last one killed with a full team of players who didn't even break 100 damage I should not suffer a PSR loss. Each player can only carry their team so much before even the BEST PLAYER OF MWO EVER gets dragged down by poor players.

Now if I Pilot said Atlas to a resounding victory and do anywhere between 250-1200 damage the PSR should go Up. But if it is a fluke (not within normal match score for me) it should not give me a huge boost to PSR. (I don't know if the current PSR system does this or not but it should not)

As it is right now a New player can drop, do nothing, be lucky enough to win and have his match score go up. On the flip side A solid T3 player Can carry a team and still loose and Loose PSR Because the WIN/LOSS means more than individual performance in a match, unless you crank out more than 500 damage. Damage is still weighted too much in the calculation of match score. Which is what is the 2nd highest factor in determining PSR gain or loss.

Overall there are several things that affect how well a person does in a match which PGI currently does not factor into Match Score and some which they will never be able to. But weighting the PSR gain/loss by win/loss over other factors is a problem.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users